Chapter 13

The Role of Technology in Teaching

and Learning Statistics

Dave Pratt, Neville Davies, and Doreen Connor

Abstract  In this chapter the merits, or otherwise, of using technology in teaching

and learning statistics are considered. The many affordances that technological
advances offer to teachers of statistics and the issues that hinder their widespread
use in classrooms are summarised. When statisticians do statistics they get involved
with far deeper concepts and carry out activities that require a wider range of
cognitive skills compared with just applying techniques. It seems that pedagogic
developments have not kept pace with those in software design, in that the
opportunity to use computers to engage students in the full statistical enquiry cycle
is not being exploited. The authors believe that beginning teachers must be exposed
to such opportunities if they are to appreciate the key role that technology could
have in facilitating the development of students’ understanding of statistics.

1 Introduction

We are all familiar with the increasing importance of technology within all realms
of our existence but what role does technology have in the teaching and learning
of statistics? In responding to this question, this chapter focuses on the potential of

D. Pratt (*)
Institute of Education, University of London, 20 Bedford Way, London WC1H 0AL, UK
e-mail: [email protected]
N. Davies
Royal Statistical Society Centre for Statistical Education,
University of Plymouth, Plymouth, UK
e-mail: [email protected]
D. Connor
Clifton Campus, Nottingham, NG11 8NS, UK
e-mail: [email protected]

C. Batanero, G. Burrill, and C. Reading (eds.), Teaching Statistics in School 97

Mathematics-Challenges for Teaching and Teacher Education: A Joint ICMI/IASE Study,
DOI 10.1007/978-94-007-1131-0_13, © Springer Science+Business Media B.V. 2011
98 D. Pratt et al.

technology in the teaching and learning of statistics, recognising however that those
potentials are mediated inevitably by many factors, including inter alia software,
tasks and teaching approaches. Harradine (2008) has noted how in the late 1990s
statistical investigations began to incorporate real and large data sets as computers
managed the number crunching. However, he has expressed concern that even now
pedagogy in statistics education appears to have not moved on. This chapter
explores some of the opportunities technology offers teachers of statistics while
remaining alert to the issues surrounding their use. The authors’ emphasis is similar
to that of Goldstein (2003), namely, that it is not good enough to only consider
which technology to use, but that, in order for effective learning to take place, it is
how the technology is integrated into the curriculum and learning process and how
the teacher uses it that are vital. In this chapter, we focus both on the use of
computers themselves and some specific software tools.

2 What Does Technology Offer Teachers of Statistics?

This section discusses the special opportunities for teaching statistics that technology
offers teachers who aim to provide rich learning experiences for their students.
Ben-Zvi (2000) proposed the following categories of software: statistical packages
(tools), microworlds, tutorials, resources (including Internet resources) and teachers’
metatools. The discussion below considers five important affordances for teaching
that might accrue should these categories of software be more widely adopted.

2.1 Using Representations as Dynamic Tools for Analysis

Traditionally, graphs are used to report data, often through displays and presentations.
Exploratory Data Analysis (EDA) encourages deep interaction with data typically
supported by the sort of immediate graphical representation of data that can be
generated by computers. Computers thus enable representations to be used as
analytical tools during an investigation rather than only as presentational tools
at the end of the investigation. When graphs are used to try to make sense of
data during analysis, the representations need to appeal to an intuitive sense
of position, spread and outlying values. One example is the hat plot in Tinkerplots
(www.keypress.com/x5715.xml), designed to appeal to students’ intuitive notion
of a modal clump (Konold et  al., 2002). Similarly, Cobb, Gravemeijer, Bowers,
and McClain (1997) developed their mini-tools as part of an intuitive infrastructure
within a learning trajectory for statistical ideas based around the affordance of
computer software to enable dynamic manipulation of images and numerical
data. McClain (2008) reported teacher use of these mini-tools to develop ways of
thinking about distribution.
There are forms of representation that are as yet underdeveloped. For example,
the manipulation of multivariate data might be better supported through digital
13  Technology in Teaching and Learning Statistics 99

technology. Ridgway, Nicholson, and McCusker (2008) reported on the development

of new web browser tools (www.dur.ac.uk/smart.centre/) for displaying up to six
variables for the easy exploration of complex data sets.

2.2 Expressing Personal Models

Statistical analysis involves the creation and use of models that describe the data
arising from the phenomenon in question. Teachers of statistics might therefore
involve their students in the activity of expressing personal models that attempt to
capture the inherent structure in a situation. In this way, students might embrace the
full cycle of statistical enquiry (discussed in more detail later) through the stages of
reality description and pseudo-concrete model, mathematisation-formalisation and
validation (Henry, Girard, & Chaput, 2008).
Computers offer flexible tools that empower the levels of expressiveness needed
to develop models that fit data. In EDA, students express their own informal models
for the data by searching for trends and patterns in the data, a process often referred
to as expressive modelling (Doerr & Pratt, 2008). New developments in Tinkerplots
promise to provide a graphical probabilistic language to model the generation of
data sets (Konold, Harradine, & Kasak, 2007). Teachers could use the software as
an authoring tool in which they build models for students to explore or as an
expressive tool in which students build their own models of phenomena.
It is only by engaging with models that students might become tuned towards the
uncertainty in the model (Chatfield, 1995). There are opportunities here to compare
the logical necessity of mathematics and the vagaries of statistics. Biehler (2008),
for example, has argued that the automatic graph plotting, now available through
com­puters and graphing calculators fitted with data logging devices, offers a
natural platform for discussing idealised mathematical functions alongside pro­cesses
con­taining noise as in statistical situations. Early work in this area was done by
Ainley, Pratt, and Nardi (2001), who proposed a pedagogic technique called active
graphing, in which students carried out experiments to generate noisy data.
Comparison between the data arising from the manual data collection and the smooth
mathematical functions arising from digital representations raises the issue for
discussion about the difference between mathematical and statistical data and the
role of mathematical functions as models underpinning data. However, in teaching
statistics we should always bear in mind that, in building models for data, the
insightful comment of Box (1979) that all models are wrong, but some are useful
is highly relevant.

2.3 Exploring Models

There are a number of key statistical concepts with which teachers expect students to
engage. Expressive modelling might provide opportunities for students to appreciate
the utility of those concepts in specific contexts but expressive modelling is
100 D. Pratt et al.

unpre­dictable and cannot guarantee engagement with any specific concept. An

alternative is for models to be built into the computer software and used to generate
simulations that can be explored by the student, who might see the real-world
phenomenon through the mathematical model (rather than see the model through
the data). Integrating simulation into teaching can have positive benefits by allowing
students to experiment with data and statistical distributions (Garfield & Ben-Zvi,
2007, 2008; Engel, Sedlmeier, & Worn, 2008). Such approaches are referred to
as exploratory modelling (Doerr & Pratt, 2008). The computational power is
directed towards providing feedback according to the in-built model in response
to the action of the student.
Simulations have been used to help students bridge ambiguous statistical concepts.
Some examples are Engel et  al. (2008) on explained and unexplained variation;
Kadijevich, Kool-Voljic, and Lavicza (2008) on sampling distributions; Abrahamson
and Wilensky (2007) on different epistemological perspectives for probability;
Prodromou and Pratt (2006) on determined and stochastic causality; and the Winton
programme for the Public Understanding of Risk (www.understandinguncertainty.org)
on absolute and relative risk.

2.4 Storing and Processing Real Data

Digital technology facilitates the use of large data sets through its capacity for
data storage, easy retrieval and universal availability thanks to the increasing
use of idealised data formats. Data sets enable the analysis of data drawn from
situations that are meaningful to students.
One of the most significant developments for schools in this respect has been
the development of CensusAtSchool (Connor, Davies, & Holmes, 2000; Davies,
Connor, & Spencer, 2003). On the Royal Statistical Society Centre for Statistical
Education (RSSCSE) web site (www.censusatschool.org.uk), there are over 30
databases containing 1.3 million responses of real data, collected from learners in
five countries, available for sampling.
Hall (2008) described her use of CensusAtSchool with elementary teachers. She
noted that although the data sets were powerful, it was equally important to have
available a dynamic statistical analysis software, such as Tinkerplots. Web-based tools
that allow the intuitive graphing of data sets, such as the tools in CensusAtSchool
(www.censusatschool.org.uk/get-data/datatool), are increasing in availability.
Real data sets present issues that are often not present in sanitised data. For
example, difficult numbers, errors in data and missing values are all qualities of
data that might be avoided in carefully prepared situations. At some point in a
student’s education, these issues need to be confronted since they raise important
questions about the limitations, scope and reliability of inferences that can be made,
as well as techniques for handling the problems (for further discussion of using real
data in statistics education, see Hall, this book).
13  Technology in Teaching and Learning Statistics 101

2.5 Sharing and Communicating

Ben-Zvi (2007) has argued for the creation of statistics courses using a wiki to
promote collaborative learning. He has researched several types of wiki activities:
collaborative writing; glossaries; discussion and review; statistical projects;
self-reflective journals; and assessment. The creation of such resources could be
beneficial to teacher trainees, both for their own learning and for the future benefit
of their students.
Nolan and Temple Lang (2007) showed how a large meaningful data set
(9,000 email messages containing 30 variables for each message) was used to teach
statistical practice in a dynamic document environment. Actions allowed by the live
worksheet include interacting with the data, modifying inputs, changing computa­tions
and exploring a range of analyses. By allowing the electronic docu­ment to be
interactive, the authors demonstrated how it could be used to reflect what a statistician
might do in carrying out a statistical investigation. In essence they aimed to provide
a problem-solving environment in which practising statisticians and statistical
researchers work so that educators/teachers could more effectively teach students.

3 Issues Regarding the Use of Technology in Teaching

and Learning Statistics

It is true that technology has changed the way people consume statistics, the speed
and efficiency with which researchers produce statistics and how statisticians do
statistics. We might also expect the power of technology as described above to
change the way people teach statistics and students learn statistics (Chance,
Ben-Zvi, Garfield, & Medina, 2007). Rubin (2007) reflected on experience in using
technology in statistics education between 1992 and 2007 and commented that “…
as amazing and inspiring as these technologies may seem, none of them have any
educational effect without carefully constructed curriculum and talented teaching”
(p. 2). In fact, teaching and learning is such a complex process that the very power
that creates potential for dramatic change comes hand-in-hand with a set of issues
that threaten to curtail the possible development. Five of these issues are elaborated.

3.1 Teachers May Not Prioritise the Use of Technology

In the end, what matters is what students learn but there is little statistics education
research about teachers using such technological tools (Chance et  al., 2007).
In prac­tice, teachers have to make decisions about what is appropriate to incorporate
into already-written curricula and have little time to research such matters. They
also have competing demands on classroom time and are forced to prioritise. Placing
102 D. Pratt et al.

more emphasis on enquiry and investigation, and using new forms of technology
appear, on the face of it, to be time-consuming and so demand a commitment to a
pedagogical approach that many teachers may not share.

3.2 Curriculum Specifications May Not Support

the Use of Technology

Teachers’ priorities might be influenced by curriculum specification. However,

there is much variation in the use of technology between the statistics curricula
that schools follow within and between countries. In Australia, for example, all
curriculum documents include emphasis on the use of technology, but the extent to
which this enables statistical data investigations depends on the assessment regime.
A common technology used in Australia is the graphics calculator, but the flexibility
in the Queensland system is seeing more schools making use of a combination of
computers and graphics calculators than in other states.

3.3 Assessment Methods May Not Encourage

the Use of Technology

Even if the curriculum encourages the use of computers, teachers realise that
examinations may be more important to their students and the students’ parents than
anything else. The actual curriculum as experienced by students will always be driven
by the assessment regime, irrespective of pedagogic guidelines. In many countries
it remains the case that examination boards do not allow the use of technology.
For example, in Queensland, the assessment for senior school is school-based
cen­trally moderated assessment; it is mandatory in mathematics (including statistics)
to include alternative assessments such as investigations. In contrast, the emphasis
in New South Wales is almost entirely on a final statewide exam and this has
probably contributed to the lack of progress in developing students’ statistical
conceptual understanding. Teachers who consider the use of technology in their
classrooms may be concerned that their students will not develop practices that
transfer to such traditional examination contexts.

3.4 Teachers May Not Re-skill

Experienced teachers may not have benefited from the use of computers in their
own learning. Their own experience of being taught is unlikely to have included the
changes in pedagogy that Chance et al. (2007) have claimed flow from techno­logical
advances, including balancing the role of computers and non-technological
13  Technology in Teaching and Learning Statistics 103

environments for teaching how to “unlock stories in data” (Pfannkuch, 2008). Garfield
and Ben-Zvi (2007) have supported exposing teachers to innovative software,
such as TinkerPlots and Fathom, to enable them to explore data. The intention is
that exposure during teacher education might motivate teachers to pass their
experience on and engage their students in statistical investigations through using
similar software.

3.5 Technology May Not Be Used to Teach Statistical Concepts

but Instead May Reinforce Emphasis on Computation

Technology as a productivity tool has been concerned with developing faster and
faster ways of processing data to calculate routine and/or sophisticated statistics and,
more recently, with processing very large databases. However, the way statisticians
do statistics also involves planning the investigation, collecting data, processing,
analysing results and drawing appropriate conclusions (Marriott, Davies, & Gibson,
2009; Stuart, 1995). Therefore it is important to stress that, when teaching the
subject, processing and analysing results is just one part of the statistical enquiry
cycle. The way that computers can support an emphasis on modelling as well as on
analysis, as discussed above, could ensure that students do not lose touch with the
statistical thinking necessary to do statistics when they use technology.

4 Challenges for Teacher Education

It has been argued that technology offers a range of affordances that could revolu­tionise
statistics education but there are reasons why that transformation has not yet
happened and indeed might never happen, except in certain pockets of good
practice. For a number of years, statistics educators have proposed that the way
statistics is taught should be changed (Stuart, 1995; Chance et al., 2007; Marriot
et al., 2009), and that the place to begin is in teacher education.

4.1 Immersing Beginning Teachers in the Use of Technology

for Teaching Statistics

Da Ponte (2008) commented on the need for a better vision of how teachers learn.
Studies have reported on the difficulties teachers face due to their own lack of
exposure and knowledge of the best use of technology (Reston & Bersales, 2008)
and yet beginning teachers are often heavily influenced in their teaching by the way
that they themselves were taught (Ball, 1988). When new teachers teach areas in
104 D. Pratt et al.

which their own conceptual knowledge is weak, they have been shown to revert more
readily to those methods they themselves experienced as a learner (Sedlmeier &
Wassner, 2008). The use of a wide range of technology can present problems for
beginning teachers, who are accustomed to a certain style of instruction (Healy &
Hoyles, 2001). Immersion in the use of technology for statistics education should
therefore be a major feature of any teacher education course. However, teacher
educators should first explore the many ideas being put forward by researchers and
developers (Batanero, Godino, & Roa, 2004).

4.2 How Might Teacher Educationalists Deploy Technology?

Immersion, though, must involve the full statistical enquiry cycle as discussed
above. Using modern software to express and explore models of real data sets,
such as those available through CensusAtSchool, would expose weaknesses in the
beginning teachers’ own statistical knowledge in a non-threatening environment.
Since Lee and Hollebrands (2008) reported that teachers’ decisions about using
computer tools were often based on knowledge gained during their teacher educa­tion
courses, it is reasonable to suppose that a modelling approach in teacher education
courses might enable the development of a more sophisticated understanding from
the exposed weaknesses. This immersion in the use of statistics software in teacher
education courses is of vital importance due to the gearing effect on eventual
student learning of statistical ideas and because teacher educators have potential
influence over many students who later become teachers.

4.3 What Might Beginning Teachers Learn About Statistics

Through Using Technology?

The aim of such immersion in teacher education would be to enhance teachers’

technological pedagogical content knowledge (TPCK) (Lee & Hollebrands, 2008),
distinguishing between technology as course content and technology as a teaching
tool (Habre & Grundmeier, 2007). Learning how the special characteristics of
techno­logy might be integrated into teaching to support learning of statistical
concepts can be considered a key aspect of TPCK. Pfannkuch (2008) stated that
teacher educators need both to build teachers’ statistical concepts and to make
teachers aware of how students’ conceptual understanding may develop.
The focus on EDA, supported by technology, can aid beginning teachers
themselves engage with statistical concepts and the whole statistical investigation
cycle, helping them to reach a higher level of conceptual understanding of statistics
before considering how they might use technology in their own teaching, thus also
enabling students to understand statistical concepts.
13  Technology in Teaching and Learning Statistics 105

5 The Way Forward

The development of technologies, especially in the form of new software tools, will
of course continue and no doubt so will the careful analysis of how their design and
implementation impacts on teaching and learning. Nevertheless, the way forward is
to place emphasis on teacher education so that the affordances of technology,
as identified in Sect. 2 of this chapter, can be understood by teachers, the issues
alerted in Sect. 3 can be shared and avoided, and finally so that the challenges in
the final section above can be met. Research is needed to examine and evaluate
such developments in teacher education and to help understand and measure
the presumed consequent gearing effect that results from the way that teachers
influence many students in their courses.

References

Abrahamson, D., & Wilensky, U. (2007). Learning axes and bridging tools. International Journal
of Computers for Mathematics Learning, 12(1), 23–55.
Ainley, J., Pratt, D., & Nardi, E. (2001). Normalising: children’s activity to construct meanings for
trend. Educational Studies in Mathematics, 45, 131–146.
Ball, D. L. (1988). Unlearning to teach mathematics. (Issue Paper 88–1). East Lansing: Michigan
State University, National Center for Research on Teacher Education.
Batanero, C., Burrill, G., Reading, C., & Rossman, A. (Eds.) (2008). Joint ICMI/IASE Study:
Teaching Statistics in School Mathematics. Challenges for Teaching and Teacher Education.
Proceedings of the ICMI Study 18 and 2008 IASE Round Table Conference. Monterrey,
Mexico: International Commission on Mathematical Instruction and International Association
for Statistical Education. Online: www.stat.auckland.ac.nz/~iase/publications
Batanero, C., Godino, J., & Roa, R. (2004). Training teachers to teach probability. Journal of
Statistics Education, 12(1). Online: www.amstat.org/publications/jse/
Ben-Zvi, D. (2000). Toward understanding the role of technological tools in statistical learning.
Mathematical Thinking and Learning, 2(1 & 2), 127–155.
Ben-Zvi, D. (2007). Using wiki to promote collaborative learning in statistics education. Technology
Innovations in Statistics Education, 1(4). Online: repositories.cdlib.org/uclastat/cts/tise/
Biehler, R. (2008). From statistical literacy to fundamental ideas in mathematics: How can
we bridge the tension in order to support teachers of statistics. In C. Batanero, G. Burrill,
C. Reading, & A. Rossman (2008).
Box, G. E. P. (1979). Robustness in the strategy of scientific model building. In R. L. Launer &
G. N. Wilkinson (Eds.), Robustness in statistics. New York: Academic Press.
Chance, B., Ben-Zvi, D., Garfield, J., & Medina, E. (2007). The role of technology in improving
student learning in statistics. Technology Innovations in Statistics Education, 1(1). Online:
repositories.cdlib.org/uclastat/cts/tise/
Chatfield, C. (1995). Model uncertainty, data mining and statistical inference. Journal of the Royal
Statistical Society A, 158, 419–466.
Cobb, P., Gravemeijer, K. P. E., Bowers, J., & McClain, K. (1997). Statistical minitools. Nashville,
TN/Utrecht: Vanderbilt University and Freudenthal Institute/Utrecht University.
Connor, D., Davies, N., & Holmes, P. (2000). CensusAtSchool 2000. Teaching Statistics,
22, 66–70.
da Ponte, J. P. (2008). Preparing teachers to meet the challenges of statistics education.
In C. Batanero, G. Burrill, C. Reading, & A. Rossman (2008).
106 D. Pratt et al.

Davies, N., Connor, D., & Spencer, N. M. (2003). An international project for the development of
data handling skills of teachers and pupils. Journal of Applied Mathematics and Decision
Sciences, 7, 75–83.
Doerr, H., & Pratt, D. (2008). The learning of mathematics and mathematical modeling. In
M. K. Heid & G. W. Blume (Eds.), Research on technology in the teaching and learning of
mathematics: Syntheses and perspectives. Mathematics learning, teaching and policy (Vol. 1,
pp. 259–285). Charlotte, NC: Information Age.
Engel, J., Sedlmeier, P., & Worn, C. (2008). Modelling scatter plot data and the signal-noise
metaphor: towards statistical literacy for pre-service teachers. In C. Batanero, G. Burrill,
C. Reading, & A. Rossman (2008).
Garfield, J., & Ben-Zvi, D. (2007). How students learn statistics revisited: A current review of
research on teaching and learning statistics. International Statistical Review, 75(3), 372–396.
Garfield, J., & Ben-Zvi, D. (2008). Preparing schoolteachers to develop students’ statistical
reasoning. In C. Batanero, G. Burrill, C. Reading, & A. Rossman (2008).
Goldstein, R. (2003). Integrating computers into the teaching of mathematics. In L. Haggarty
(Ed.), Teaching mathematics in secondary schools – A reader (pp. 143–159). Milton Keynes:
Open University.
Habre, S., & Grundmeier, T. (2007). Prospective mathematics teachers’ views on the role of
technology in mathematics education. Issues in the Undergraduate Mathematics Preparation
of School Teachers: The Journal, 3. Online: www.k-12prep.math.ttu.edu
Hall, J. (2008). Using CensusAtSchool and Tinkerplots to support Ontario elementary teachers’
statistics teaching and learning. In C. Batanero, G. Burrill, C. Reading, & A. Rossman (2008).
Harradine, A. (2008). Birthing big ideas in the minds of babes. In C. Batanero, G. Burrill, C. Reading,
& A. Rossman (2008).
Healy, L., & Hoyles, C. (2001). Software tools for geometrical problem solving: Potentials and
pitfalls. International Journal of Computers for Mathematical Learning, 6(3), 235–256.
Henry, M., Girard, J., & Chaput, B. (2008). Modelling and simulations in statistics education.
In C. Batanero, G. Burrill, C. Reading, & A. Rossman (2008).
Kadijevich, D., Kool-Voljic, V., & Lavicza, Z. (2008). Towards a suitably designed instruction
on statistical reasoning: Understanding sampling distribution with technology. In C. Batanero,
G. Burrill, C. Reading, & A. Rossman (2008).
Konold, C., Harradine, A., & Kasak, S. (2007). Understanding distributions by modeling them.
International Journal of Computers for Mathematical Learning, 12(3), 217–230.
Konold, C., Robinson, A., Khalil, K., Pollatsek, A., Well, A., Wing, R., & Mayr, S. (2002).
Students’ use of modal clumps to summarize data. In B. Phillips (Ed.), Proceedings of the
Sixth International Conference on Teaching Statistics. Cape Town, South Africa: International
Association for Statistics Education. Online: www.stat.auckland.ac.nz/~iase/publications
Lee, S., & Hollebrands, K. (2008). Preparing to teach data analysis and probability with
technology. In C. Batanero, G. Burrill, C. Reading, & A. Rossman (2008).
Marriott, J. M., Davies, N., & Gibson, E. (2009). Teaching, learning and assessing statistical problem
solving. Journal of Statistics Education, 17(1). Online: www.amstat.org/publications/jse/
McClain, K. (2008). The evolution of teachers’ understandings of distribution. In C. Batanero,
G. Burrill, C. Reading, & A. Rossman (2008).
Nolan, D., & Temple Lang, D. (2007). Dynamic, interactive documents for teaching statistical
practice. International Statistical Review, 75(3), 295–321.
Pfannkuch, M. (2008). Training teachers to develop statistical thinking. In C. Batanero, G. Burrill,
C. Reading, & A. Rossman (2008).
Prodromou, T., & Pratt, D. (2006). The role of causality in the coordination of two perspectives
on distribution within a virtual simulation. Statistics Education Research Journal, 5(2), 69–88.
Online: www.stat.auckland.ac.nz/serj/
Reston, E., & Bersales, L. G. (2008). Reform efforts in training statistics teachers in the Philippines:
Challenges and prospects. In C. Batanero, G. Burrill, C. Reading, & A. Rossman (2008).
Ridgway, J. Nicholson, J., & McCusker, S. (2008). Reconceptualising ‘statistics’ and ‘education’.
In C. Batanero, G. Burrill, C. Reading, & A. Rossman (2008).
13  Technology in Teaching and Learning Statistics 107

Rubin, A. (2007). Much has changed; little has changed: Revisiting the role of technology in
statistics education 1992–2007. Technology Innovations in Statistics Education, 1(1). Online:
repositories.cdlib.org/uclastat/cts/tise/
Sedlmeier, P., & Wassner, C. (2008). German mathematics teachers’ views on statistics education.
In C. Batanero, G. Burrill, C. Reading, & A. Rossman (2008).
Stuart, M. (1995). Changing the teaching of statistics. The Statistician, 44, 45–54.